Hierarchical Structure Generation for Malayalam using Hybrid Approach

ABSTRACT In this paper, a hierarchical structure is generated for each Malayalam sentence that contains all grammatical information including subject?object and gender information. This structure resembles to tree structure, root ‘S’ represents sentence and each branch represents information of each word in the sentence and leaf nodes are terminals that are only Malayalam words. This detailed syntactical structure including gender is developed and tested against a corpus of about 100 sentences (740 words), which are entered in Malayalam script.